PROJECT SUMMARY/ABSTRACT Diffuse large B cell lymphoma (DLBCL) compromises nearly 40% of all newly diagnosed non-Hodgkin lymphomas (NHL) in the U.S. Currently, mainly clinical factors are used to identify high versus low risk patients and to predict overall survival, but with modest utility. We have recently shown that newly diagnosed patients treated with anthracycline-based immunochemotherapy (standard of care) who do not have disease progression or relapse, retreatment, or death within 24 months of diagnosis (termed event-free survival or EFS24) for DLBCL have a normal life expectancy, while patients who fail to achieve this sentinel landmark have very poor outcomes. While we have identified clinical predictors of these endpoints, we hypothesize that both somatic (tumor) and germline (host) genetic biomarkers can improve our ability to predict at diagnosis which patients will have an early clinical failure and help identify novel biology and therapeutic approaches. Our preliminary data from whole- exome sequencing and genome-wide association studies in DLBCL patients suggests that there are both somatic and germline genetic biomarkers that predict failure to achieve EFS24, and that these are different from known drivers and susceptibility loci. Building on these findings, we propose the following specific aims are: 1) To identify, validate and clinically translate somatic genetic biomarkers for failure to achieve EFS24; 2) To identify, validate, and clinically translate germline genetic biomarkers for failure to achieve EFS24; and 3) To develop and validate an integrative model that combines somatic and germline genetic biomarkers with clinical prognostic factors. For each aim, we will also initially assess biomarker function through the use of bioinformatics tools, and when applicable, with initial laboratory interrogation in model systems. We will address our aims using the established and extensive resources of the Mayo Clinic and University of Iowa Lymphoma Specialized Program of Research Excellence (SPORE) and Lymphoma Epidemiology of Outcomes (LEO) cohort, and established collaborations with the LYSA (French Lymphoma Trials Group) and the Mayo Clinic Department of Laboratory Medicine and Pathology. We anticipate that our results will have a major impact on the management of DLBCL by leading to clinical translation of both host and tumor genetic biomarkers for risk stratification, identifying high-risk patients who might benefit from more intensive therapies or non-standard first-line drug regimens, and identification of novel biology and therapeutic targets.